Update app.py
Browse filesSupport multi channel
better ASR model
app.py
CHANGED
@@ -1,4 +1,3 @@
|
|
1 |
-
|
2 |
from random import sample
|
3 |
import gradio as gr
|
4 |
import torchaudio
|
@@ -10,8 +9,12 @@ import jiwer
|
|
10 |
|
11 |
# ASR part
|
12 |
from transformers import pipeline
|
13 |
-
p = pipeline("automatic-speech-recognition")
|
14 |
-
|
|
|
|
|
|
|
|
|
15 |
# WER part
|
16 |
transformation = jiwer.Compose([
|
17 |
jiwer.ToLowerCase(),
|
@@ -21,10 +24,10 @@ transformation = jiwer.Compose([
|
|
21 |
])
|
22 |
|
23 |
# WPM part
|
24 |
-
from transformers import
|
25 |
processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-xlsr-53-espeak-cv-ft")
|
26 |
phoneme_model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-xlsr-53-espeak-cv-ft")
|
27 |
-
|
28 |
class ChangeSampleRate(nn.Module):
|
29 |
def __init__(self, input_rate: int, output_rate: int):
|
30 |
super().__init__()
|
@@ -44,7 +47,9 @@ class ChangeSampleRate(nn.Module):
|
|
44 |
model = lightning_module.BaselineLightningModule.load_from_checkpoint("epoch=3-step=7459.ckpt").eval()
|
45 |
|
46 |
def calc_mos(audio_path, ref):
|
47 |
-
wav, sr = torchaudio.load(audio_path, channels_first=True)
|
|
|
|
|
48 |
osr = 16_000
|
49 |
batch = wav.unsqueeze(0).repeat(10, 1, 1)
|
50 |
csr = ChangeSampleRate(sr, osr)
|
@@ -73,6 +78,7 @@ def calc_mos(audio_path, ref):
|
|
73 |
|
74 |
return predic_mos, trans, wer, phone_transcription, ppm
|
75 |
|
|
|
76 |
description ="""
|
77 |
MOS prediction demo using UTMOS-strong w/o phoneme encoder model, which is trained on the main track dataset.
|
78 |
This demo only accepts .wav format. Best at 16 kHz sampling rate.
|
@@ -83,15 +89,16 @@ Add ASR based on wav2vec-960, currently only English available.
|
|
83 |
Add WER interface.
|
84 |
"""
|
85 |
|
|
|
86 |
iface = gr.Interface(
|
87 |
fn=calc_mos,
|
88 |
inputs=[gr.Audio(type='filepath', label="Audio to evaluate"),
|
89 |
-
gr.Textbox(placeholder="Input
|
90 |
-
outputs=[gr.Textbox(placeholder="
|
91 |
gr.Textbox(placeholder="Hypothesis", label="Hypothesis"),
|
92 |
-
gr.Textbox(placeholder="Word Error Rate", label = "WER"),
|
93 |
gr.Textbox(placeholder="Predicted Phonemes", label="Predicted Phonemes"),
|
94 |
-
gr.Textbox(placeholder="Phonemes per minutes", label="PPM")],
|
95 |
title="Laronix's Voice Quality Checking System Demo",
|
96 |
description=description,
|
97 |
allow_flagging="auto",
|
|
|
|
|
1 |
from random import sample
|
2 |
import gradio as gr
|
3 |
import torchaudio
|
|
|
9 |
|
10 |
# ASR part
|
11 |
from transformers import pipeline
|
12 |
+
# p = pipeline("automatic-speech-recognition")
|
13 |
+
p = pipeline(
|
14 |
+
"automatic-speech-recognition",
|
15 |
+
model="KevinGeng/whipser_medium_en_PAL300_step25",
|
16 |
+
device=0,
|
17 |
+
)
|
18 |
# WER part
|
19 |
transformation = jiwer.Compose([
|
20 |
jiwer.ToLowerCase(),
|
|
|
24 |
])
|
25 |
|
26 |
# WPM part
|
27 |
+
from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
|
28 |
processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-xlsr-53-espeak-cv-ft")
|
29 |
phoneme_model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-xlsr-53-espeak-cv-ft")
|
30 |
+
# phoneme_model = pipeline(model="facebook/wav2vec2-xlsr-53-espeak-cv-ft")
|
31 |
class ChangeSampleRate(nn.Module):
|
32 |
def __init__(self, input_rate: int, output_rate: int):
|
33 |
super().__init__()
|
|
|
47 |
model = lightning_module.BaselineLightningModule.load_from_checkpoint("epoch=3-step=7459.ckpt").eval()
|
48 |
|
49 |
def calc_mos(audio_path, ref):
|
50 |
+
wav, sr = torchaudio.load(audio_path, channels_first=True)
|
51 |
+
if wav.shape[0] > 1:
|
52 |
+
wav = wav.mean(dim=0, keepdim=True) # Mono channel
|
53 |
osr = 16_000
|
54 |
batch = wav.unsqueeze(0).repeat(10, 1, 1)
|
55 |
csr = ChangeSampleRate(sr, osr)
|
|
|
78 |
|
79 |
return predic_mos, trans, wer, phone_transcription, ppm
|
80 |
|
81 |
+
|
82 |
description ="""
|
83 |
MOS prediction demo using UTMOS-strong w/o phoneme encoder model, which is trained on the main track dataset.
|
84 |
This demo only accepts .wav format. Best at 16 kHz sampling rate.
|
|
|
89 |
Add WER interface.
|
90 |
"""
|
91 |
|
92 |
+
|
93 |
iface = gr.Interface(
|
94 |
fn=calc_mos,
|
95 |
inputs=[gr.Audio(type='filepath', label="Audio to evaluate"),
|
96 |
+
gr.Textbox(placeholder="Input reference here (Don't keep this empty)", label="Reference")],
|
97 |
+
outputs=[gr.Textbox(placeholder="Naturalness evaluation, ranged 1 to 5, the higher the better.", label="Predicted MOS"),
|
98 |
gr.Textbox(placeholder="Hypothesis", label="Hypothesis"),
|
99 |
+
gr.Textbox(placeholder="Word Error Rate: Only valid when Reference is given", label = "WER"),
|
100 |
gr.Textbox(placeholder="Predicted Phonemes", label="Predicted Phonemes"),
|
101 |
+
gr.Textbox(placeholder="Speaking Rate, Phonemes per minutes", label="PPM")],
|
102 |
title="Laronix's Voice Quality Checking System Demo",
|
103 |
description=description,
|
104 |
allow_flagging="auto",
|